/sprint-news

Jira Cloud add-on to communicate sprint goals, benefits and roadmaps to stakeholders.

Primary LanguageJavaGNU General Public License v3.0GPL-3.0

Build Status

SprintNews Jira add-on

A plugin to generate a report with content tailored for stakeholders, with the goal of communicating the benefits stakeholders will get and managing their expectations on the product development. It highlights the stakeholders addressed in the selected sprint, and the value they will get after stories get implemented. Up to five roadmaps are summarized, as well as previous sprints. You can visit the plugin on Jira's Market place using this link

Sample report generated using the plugin:

Sample report generated using the plugin

Text generation algorithm

The text generation algorithm tries to adapt the request and benefit inside the user story into a text describing requests and benefits for a given user. For example, for story "As a Warehouse Manager, I want to see a calendar containing the loads expected in this week, so that I can plan stock locations and make sure there are enough people to handle them" the following user benefit text will be generated: "Warehouse Manager can be sure that they can plan stock locations and make sure there are enough people to handle them with story (...)". To accomplish this there is an issue of linking these 2 parts: the text of the story and the text describing the request or benefit for a given user. Since the request / benefit text is already limited to 2 or 3 sentences, the solution was to limit the "variations" of the text extracted from the stories by analyzing their part of speech (PoS). This task was done using Apache OpenNLP. The last part was to build the block of text that will be used by the final report, which includes adding, updating and removing part of the text (tokens). To make these changes more maintainable, each individual operation was implemented as a separate rule.

Text generation algorithm

Running the project locally and installing the add-on

  1. Install ngrok and run ngrok http 8080 (remember that ngrok now requests user signup first)
  2. Copy the https address from ngrok and update parameter 'base-url' on the src/main/java/resources/application.yml file
  3. Run mvn spring-boot:run
  4. Install the add-on on Jira (menu Apps|Manage Apps, Upload button - first you have to enable Development mode using the Settings link on the same page)
  5. Paste the https url you copied from ngrok
  6. Go to one of your Jira project's backlog and you should see the "Get newspaper" button

Configurations

Environment variable SPRING_PROFILES_ACTIVE is set to production on Heroku
Uses Liquibase for database migrations